Can the Momentum Strategy be Improved by Adding a Risk Measure?

Abstract: In this thesis it is tested if it is possible to gain momentum effects on a different type of evaluation and how momentum is affected when historical risk patterns are taken into account on Swedish data. The purpose of the thesis is to investigate whether an investor can generate higher momentum profits with lower risk if risk is included in the evaluation process. The risk in this study is represented by the value at risk. To test the hypothesis it first has to be confirmed that it is possible to gain momentum profits in Sweden during the time period chosen, which it is. Thereafter Fama-MacBeth regressions are done to evaluate the explanatory power of momentum and value at risk on the returns in the holding period. In these tests momentum turns out to be the only significant factor. Thereafter, the basic properties of momentum and value at risk are tested and finally a test for spreads between momentum and value at risk in the independent sort is done by creating matrices consisting of returns from the double sorted portfolios. The spreads are a comparison between the winner and loser portfolios. The spreads together with the rest of our results indicates that it is possible for an investor to create a portfolio with higher returns and lower risk relative to the original momentum strategy.